The Role of Instrumental Variables in Causal Inference Based on Independence of Cause and Mechanism

Causal inference methods based on conditional independence construct Markov equivalent graphs and cannot be applied to bivariate cases. The approaches based on independence of cause and mechanism state, on the contrary, that causal discovery can be inferred for two observations. In our contribution, we pose a challenge to reconcile these two research directions. We study the role of latent variables such as latent instrumental variables and hidden common causes in the causal graphical structures. We show that methods based on the independence of cause and mechanism indirectly contain traces of the existence of the hidden instrumental variables. We derive a novel algorithm to infer causal relationships between two variables, and we validate the proposed method on simulated data and on a benchmark of cause-effect pairs. We illustrate by our experiments that the proposed approach is simple and extremely competitive in terms of empirical accuracy compared to the state-of-the-art methods.     

Wess-Zumino model in the causal approach

The Wess–Zumino model is analysed in the framework of the causal approach of Epstein–Glaser. The condition of invariance with respect to supersymmetry transformations is similar to gauge invariance in the Zürich formulation. We prove that this invariance condition can be implemented in all orders of perturbation theory, i.e. the anomalies are absent in all orders. This result is of a purely algebraic nature. We work consistently in the quantum framework based on the Bogoliubov axioms of perturbation theory, so no Grassmann variables are necessary. Received: 30 April 2001 / Revised version: 19 July 2001 / Published online: 31 August 2001     

对称旋转不变相干分布式非圆信号二维波达方向估计

在相干分布式非圆信号二维波达方向估计中,利用信号非圆特性可提升估计精度,但现有的低复杂度算法利用泰勒级数近似建立的旋转不变关系会引入额外误差.针对该问题,考虑中心对称的三维立体线阵,提出了一种基于对称旋转不变关系的二维波达方向估计算法.算法首先利用信号非圆特性建立了扩展阵列模型;然后证明了对于任意的中心对称阵列,相干分布源的确定性角信号分布函数矢量具有对称特性,利用此特性在三维立体线阵的三个子阵中分别建立了扩展广义方向矢量的对称旋转不变关系;基于此,通过无须搜索的多项式求根方式分别得到中心方位角和俯仰角估计;最后利用整个阵列广义方向矢量的对称旋转不变关系构造代价函数实现了参数匹配.理论分析和仿真实验表明,相比于现有的低复杂度算法,所提算法避免了泰勒级数近似引入的额外误差,以较小的复杂度代价获得了性能的较大提升.同时,所提算法能够实现三维空间全方位的角度估计.     

Can Special Relativity Be Derived from Galilean Mechanics Alone?

Special relativity is based on the apparent contradiction between two postulates, namely, Galilean vs. c-invariance. We show that anomalies ensue by holding the former postulate alone. In order for Galilean invariance to be consistent, it must hold not only for bodies’ motions, but also for the signals and forces they exchange. If the latter ones do not obey the Galilean version of the Velocities Addition Law, invariance is violated. If, however, they do, causal anomalies, information loss and conservation laws’ violations are bound to occur. These anomalies are largely remedied by introducing waves and fields that disobey Galilean invariance. Therefore, from these inconsistencies within classical mechanics, electromagnetism could be predicted before experiment proved its existence. Special relativity, it might be argued, would then follow naturally, either as a resolution of the resulting conflict or as an extrapolation of the path between the theories. We conclude with a review of earlier attempts to base SR on a single postulate, and point out the originality of the present work.     

Unification Theory of Different Causal Algebras and Its Applications to Theoretical Physics

This paper gives a generalization of group theory, i.e. a unification theory of different causal algebras, and its applications to theoretical physics. We propose left and right causal algebras, left and right causal decomposition algebras, causal algebra and causal decomposition algebras in terms of quantitative causal principle. The causal algebraic system of containing left (or right) identity I _jL (or I _jR ) is called as the left (or right) causal algebra, and associative law is deduced. Furthermore the applications of the new algebraic systems are given in theoretical physics, specially in the reactions of containing supersymmetric particles, we generally obtain the invariance of supersymmetric parity of multiplying property. In the reactions of particles of high energy, there may be no identity, but there are special inverse elements, which make that the relative algebra be not group, however, the causal algebra given in this paper is just a tool of severely and directly describing the real reactions of particle physics. And it is deduced that the causal decomposition algebra is equivalent to group.     

A Transformer-Based Hierarchical Variational AutoEncoder Combined Hidden Markov Model for Long Text Generation

The Variational AutoEncoder (VAE) has made significant progress in text generation, but it focused on short text (always a sentence). Long texts consist of multiple sentences. There is a particular relationship between each sentence, especially between the latent variables that control the generation of the sentences. The relationships between these latent variables help in generating continuous and logically connected long texts. There exist very few studies on the relationships between these latent variables. We proposed a method for combining the Transformer-Based Hierarchical Variational AutoEncoder and Hidden Markov Model (HT-HVAE) to learn multiple hierarchical latent variables and their relationships. This application improves long text generation. We use a hierarchical Transformer encoder to encode the long texts in order to obtain better hierarchical information of the long text. HT-HVAE’s generation network uses HMM to learn the relationship between latent variables. We also proposed a method for calculating the perplexity for the multiple hierarchical latent variable structure. The experimental results show that our model is more effective in the dataset with strong logic, alleviates the notorious posterior collapse problem, and generates more continuous and logically connected long text.     

因果代数及其在物理学中的应用

依据定量因果原理,给出了物理学中的一个因果代数的应用,当满足定量因果原理的互逆可消条件且又满足消去律的解时, 得到因果分解代数;由因果分解代数导出了结合律和单位元,进而导出了因果分解代数又具有群的结构特征,同时给出了这新代数系统在高能物理学中的应用.严格地给出了在高能物理中既不是群又不是环的反应,发现因果代数和因果分解代数是严格描述粒子物理反应的基本工具,得到了所有各种相加性、相乘性物理量和各种粒子反应都必须满足的统一恒等式,给出了因果代数和因果分解代数对高能物理的具体应用.利用因果代数的表示和超对称的R数,得到了含有超对称粒子反应中相乘性的超对称的P_R=(-1)^R对称性.还得到了一个关于电子自旋角动量的任意分量间的一个对称关系式,利用这对称关系式,可以化简多电子相互作用的计算.利用互逆可消条件定义了一般的逆元,可重新定义群,使群的公理减少一个,消除了重复定义. 关键词： 对称性 群论 因果原理 粒子物理     

An axiomatic measure of one-way quantum information

I introduce an algorithm to detect one-way quantum information between two interacting quantum systems, i.e. the direction and orientation of the information transfer in arbitrary quantum dynamics. I then build an information-theoretic quantifier of one-way information which satisfies a set of desirable axioms. In particular, it correctly evaluates whether correlation implies one-way quantum information, and when the latter is transferred between uncorrelated systems. In the classical scenario, the quantity measures information transfer between random variables. I also generalize the method to identify and rank concurrent sources of quantum information flow in many-body dynamics, enabling to reconstruct causal patterns in complex networks.     

Causal Geometry

Information geometry has offered a way to formally study the efficacy of scientific models by quantifying the impact of model parameters on the predicted effects. However, there has been little formal investigation of causation in this framework, despite causal models being a fundamental part of science and explanation. Here, we introduce causal geometry, which formalizes not only how outcomes are impacted by parameters, but also how the parameters of a model can be intervened upon. Therefore, we introduce a geometric version of “effective information”—a known measure of the informativeness of a causal relationship. We show that it is given by the matching between the space of effects and the space of interventions, in the form of their geometric congruence. Therefore, given a fixed intervention capability, an effective causal model is one that is well matched to those interventions. This is a consequence of “causal emergence,” wherein macroscopic causal relationships may carry more information than “fundamental” microscopic ones. We thus argue that a coarse-grained model may, paradoxically, be more informative than the microscopic one, especially when it better matches the scale of accessible interventions—as we illustrate on toy examples.     

Revisiting Noether's Theorem on constants of motion

In this paper we revisit Noether's theorem on the constants of motion for Lagrangian mechanical systems in the ODE case, with some new perspectives on both the theoretical and the applied side. We make full use of invariance up to a divergence, or, as we call it here, Bessel-Hagen (BH) invariance. By recognizing that the Bessel-Hagen (BH) function need not be a total time derivative, we can easily deduce nonlocal constants of motion. We prove that we can always trivialize either the time change or the BH-function, so that, in particular, BH-invariance turns out not to be more general than Noether's original invariance. We also propose a version of time change that simplifies some key formulas. Applications include Lane-Emden equation, dissipative systems, homogeneous potentials and superintegrable systems. Most notably, we give two derivations of the Laplace-Runge-Lenz vector for Kepler's problem that require space and time change only, without BH invariance, one with and one without use of the Lagrange equation.     

Causal Intuition and Delayed-Choice Experiments

The conventional explanation of delayed-choice experiments appears to violate our causal intuition at the quantum level. I reanalyze these experiments using time-reversed and time-symmetric formulations of quantum mechanics. The time-reversed formulation does not give the same experimental predictions. The time-symmetric formulation gives the same experimental predictions but actually violates our causal intuition at the quantum level. I explore the reasons why our causal intuition may be wrong at the quantum level, suggest how conventional causation might be recovered in the classical limit, propose a quantum analog to the classical block universe viewpoint, and speculate on implications of the time-symmetric formulation for cosmological boundary conditions.     

An assessment of certain causal models used in surveys on aircraft noise annoyance

The causal relationships which have so far been proposed between aircraft noise exposure, annoyance and certain “psycho-social” variables (fear of aircraft crashing, general attitude towards aviation, etc.), are re-analysed, and it is demonstrated that by using correlational analysis one can arrive at contradictory results. From the sociological surveys undertaken to date, one can derive only an ordered sequence of verbal reactions to aircraft noise, and not a causal sequence between these verbal reactions: the only clear cause of annoyance is the noise itself.It is suggested that future surveys on noise annoyance should include personality tests and health questionnaires, if one wishes to establish reliable causal sequences.     

Causal Discovery in High-Dimensional Point Process Networks with Hidden Nodes

Thanks to technological advances leading to near-continuous time observations, emerging multivariate point process data offer new opportunities for causal discovery. However, a key obstacle in achieving this goal is that many relevant processes may not be observed in practice. Naïve estimation approaches that ignore these hidden variables can generate misleading results because of the unadjusted confounding. To plug this gap, we propose a deconfounding procedure to estimate high-dimensional point process networks with only a subset of the nodes being observed. Our method allows flexible connections between the observed and unobserved processes. It also allows the number of unobserved processes to be unknown and potentially larger than the number of observed nodes. Theoretical analyses and numerical studies highlight the advantages of the proposed method in identifying causal interactions among the observed processes.     

Conformal invariance and conserved quantities of a general holonomic system with variable mass

Conformal invariance and conserved quantities of a general holonomic system with variable mass are studied. The definition and the determining equation of conformal invariance for a general holonomic system with variable mass are provided. The conformal factor expression is deduced from conformal invariance and Lie symmetry. The relationship between the conformal invariance and the Lie symmetry is discussed, and the necessary and sufficient condition under which the conformal invariance would be the Lie symmetry of the system under an infinitesimal one-parameter transformation group is deduced. The conserved quantities of the system are given. An example is given to illustrate the application of the result.     

The End of a Classical Ontology for Quantum Mechanics?

In this paper, I argue that the Shrapnel–Costa no-go theorem undermines the last remaining viability of the view that the fundamental ontology of quantum mechanics is essentially classical: that is, the view that physical reality is underpinned by objectively real, counterfactually definite, uniquely spatiotemporally defined, local, dynamical entities with determinate valued properties, and where typically ‘quantum’ behaviour emerges as a function of our own in-principle ignorance of such entities. Call this view Einstein–Bell realism. One can show that the causally symmetric local hidden variable approach to interpreting quantum theory is the most natural interpretation that follows from Einstein–Bell realism, where causal symmetry plays a significant role in circumventing the nonclassical consequences of the traditional no-go theorems. However, Shrapnel and Costa argue that exotic causal structures, such as causal symmetry, are incapable of explaining quantum behaviour as arising as a result of noncontextual ontological properties of the world. This is particularly worrying for Einstein–Bell realism and classical ontology. In the first instance, the obvious consequence of the theorem is a straightforward rejection of Einstein–Bell realism. However, more than this, I argue that, even where there looks to be a possibility of accounting for contextual ontic variables within a causally symmetric framework, the cost of such an account undermines a key advantage of causal symmetry: that accepting causal symmetry is more economical than rejecting a classical ontology. Either way, it looks like we should give up on classical ontology.     

Wavelet multiple correlation and cross-correlation: A multiscale analysis of Eurozone stock markets

Statistical studies that consider multiscale relationships among several variables use wavelet correlations and cross-correlations between pairs of variables. This procedure needs to calculate and compare a large number of wavelet statistics. The analysis can then be rather confusing and even frustrating since it may fail to indicate clearly the multiscale overall relationship that might exist among the variables. This paper presents two new statistical tools that help to determine the overall correlation for the whole multivariate set on a scale-by-scale basis. This is illustrated in the analysis of a multivariate set of daily Eurozone stock market returns during a recent period. Wavelet multiple correlation analysis reveals the existence of a nearly exact linear relationship for periods longer than the year, which can be interpreted as perfect integration of these Euro stock markets at the longest time scales. It also shows that small inconsistencies between Euro markets seem to be just short within-year discrepancies possibly due to the interaction of different agents with different trading horizons.     

On continuous causal isomorphisms on two-dimensional Minkowski spacetimes

It is shown that continuous causal isomorphisms on two-dimensional Minkowski spacetime can be characterized by the invariance of wave equations.     

Conformal invariance and conserved quantity of Hamilton systems

This paper studies conformal invariance and conserved quantities of Hamilton system. The definition and the determining equation of conformal invariance for Hamilton system are provided. The relationship between the conformal invariance and the Lie symmetry are discussed, and the necessary and sufficient condition that the conformal invariance would be the Lie symmetry of the system under the infinitesimal one-parameter transformation group is deduced. It gives the conserved quantities of the system and an example for illustration.     

Conformal invariance and conserved quantities of general holonomic systems in phase space

This paper studies the conformal invariance and conserved quantities of general holonomic systems in phase space. The definition and the determining equation of conformal invariance for general holonomic systems in phase space are provided. The conformal factor expression is deduced from conformal invariance and Lie symmetry. The relationship between the conformal invariance and the Lie symmetry is discussed, and the necessary and sufficient condition that the conformal invariance would be the Lie symmetry of the system under the infinitesimal single-parameter transformation group is deduced. The conserved quantities of the system are given. An example is given to illustrate the application of the result.